Comparison between Deterministic methods and the Artificial Bee Colony Algorithm, for the Economic Load Dispatch, turning off the Generators of Higher Cost

Abstract

The reduction of fuel costs in the production of electric power in Power Plants (PP) is one of the most significant problems in this industry. This problem is known as optimization of the Economic Load Dispatch (ELD). The objective of this paper is to analyze the application of the Artificial Bee Colony Algorithm (ABC) metaheuristics, considering the incremental cost of fuel and the shutdown of the engines with the highest fuel cost per MWh in UTEs whose installed capacity exceeds the required power demand. Several techniques have been developed to solve the problem of ELD, among them: lambda iteration method, gradient method, Newton method and so on. The results for this study with the application of ABC, considering the shutdown of the generators of higher cost per MWh, obtaining a mean reduction of 6.54% in the total fuel cost, compared to the classic solutions that use all engines of the UTE. In addition to cost reduction, this proposal helps the specialist responsible for the management of the UTE in the decision making of the preventive maintenance of the engines that are not being used at the moment of the optimization, improving not only the generation efficiency, but also the generation planning of the plant.